Download Oil Price Impacts and Multifamily Housing

May 2015
Steve Guggenmos | 571-382-3520
Harut Hovsepyan | 571-382-3143
Sara Hoffmann | 571-382-5916
Oil Price Impacts and Multifamily
Housing
As oil prices fall and are expected to remain low, energy-dependent
areas are at risk of an economic slowdown and impacts to their
multifamily markets.
 Smaller metropolitan areas will likely experience a more severe
economic impact, whereas larger metro areas are more diversified
and can better absorb the potential losses.
 Houston is the largest metro with risk exposure; while the oil price
drop will have an impact on the multifamily sector there, it will not
be as severe as during the energy crash of the 1980s.
 The broader economy is expected to benefit from lower oil prices
over the next few years.
_______________________________________________________

The broader economy is
expected to benefit from
the lower oil prices, but
areas that are heavily
energy-dependent are at
risk of economic
slowdown or downturn.
The recent drop in oil prices has raised concerns about the economies of energydependent areas across the country, especially as many market watchers forecast
prices to remain low through the end of the year. The last energy bust in the 1980s
severely impacted the economies of the “oil patch” region, causing several areas to
fall into recession. With oil prices down by more than 50 percent since November
2014, the risk that these regions might fall into recession has become real once again.
The risk largely resides in smaller, less-diversified economies of the oil patch region,
but a few larger metropolitan areas with a heavy dependency on oil-related industries
could also be impacted. However, the broader economy is expected to benefit from
the lower oil prices because household savings on gasoline consumption and reduced
operating costs in many industries are expected to surpass the losses in the oil
industry.
In the most exposed areas, labor markets could be impacted significantly, which
could lead to imbalance between supply and demand in the housing market. This is
1
the case in both large and small markets, but the imbalance for the apartment market
is likely to occur in the larger metropolitan areas due to their larger inventory of
multifamily stock. Even though the economies in the smaller metropolitan areas are
at greater risk of an economic slump, they traditionally do not have as much
multifamily stock. Therefore, the risk to the multifamily industry comes from the
larger metropolitan areas.
Houston is the largest metro that could experience greater risk exposure to its
multifamily sector due to the heavy concentration of energy-dependent jobs along
with its recent multifamily construction boom. We run a stress scenario to determine
the impact a negative market shock could have on the Houston multifamily rental
market. We conclude that Houston as a whole likely will not reach the level of stress
experienced in the early 1980s if oil prices remain low.
Section 1 – Risk Exposure
We start by identifying markets considered to have higher potential risk exposure
due to their dependency on the energy sector. Our analysis measures exposure using
multiple factors, including oil production level, rig count, job concentration in oiland gas-related industries, market industrial diversification, and recent multifamily
housing starts. We conclude that smaller, oil-dependent areas will see the largest
impact on jobs and the local economy, while larger areas can better absorb potential
job losses but have a higher inventory of multifamily stock that could be affected.
U.S. Oil Production
U.S. oil production has
increased substantially
across many of the oilproducing areas, but
most notably in North
Dakota and Texas,
during the past few
years.
Since the early 2000s, oil prices have generally been increasing (though there was a
period of significant decline when the broader economy was weak during the Great
Recession) and, until recently, remained elevated around $80-100 per barrel since
2010. This opened up opportunities for many energy companies to expand
production. With improvements in the technology to extract oil from more difficultto-drill places, the production of oil in the United States has increased from 170
million barrels per month in 2011 to 289 million barrels as of December 2014, a 70
percent increase. Although this is not the historical peak, the growth rate in
production over the past three years has been much steeper than any previous threeyear period since the 1950s, growing on average 15 percent per year. 1 The growth in
production has driven economic growth in the areas where oil production and
related industries make up a dominant sector of the local economy.
1
Production of U.S. oil peaked in October 1970 at 310 million barrels per month. Production then gradually
declined, but peaked again in January 1986, at 283 million barrels. After that, the production of U.S. oil trended
down, bottoming at 119 million barrels in September 2008.
2
While oil production in the United States is concentrated in several regions, the
growth pattern in each of these regions has varied over time. Exhibit 1 shows how
the share of U.S. oil production has changed since 1980. North Dakota is now the
third largest producer of oil in the United States as production increased by 400
percent in the 2010s, relative to average production a decade earlier. Meanwhile,
levels as well as the share of production in several other key areas have remained flat
or fallen over the past few years, including Louisiana and Alaska.
Exhibit 1 – Share of Oil Production in the United States by Decade2
Source: U.S. Energy Information Administration (EIA), Freddie Mac
Note: The majority of the wells in the Gulf of Mexico are off the coast of Louisiana, but production is shared with
Texas and Alabama
Higher-risk States
While the oil production and related industries are capital intensive, they also employ
a large number of employees. As such, swings in oil production positively or
adversely affect the labor market and overall economic well-being of areas that
depend heavily on the oil industry. Accordingly, we look at the concentration of jobs
in oil and gas production and related industries to assess which markets are at higher
risk. We also look at oil and gas rig counts in each of these markets. Changes in rig
counts can inform future production levels. The number of active rigs in the United
States has dramatically declined since the sharp drop in oil prices. The drop can
impact the current labor market, and also suggests supply will fall in the mid-term to
firm up oil prices.
Exhibit 2 captures geographic exposure to the energy sector by presenting the oil
and gas job concentration together with the share of rigs in the United States
2
While the combined share of production in states included in the “Other” group is high, share of individual states
ranges from nearly 0 to 1.8 percent.
3
Wyoming, North
Dakota and Alaska
have the highest
concentrations of oil- and
gas-related jobs, but
Texas operates the most
oil rigs.
(including land and off-shore rigs) and changes in the number of rigs over the past
few months. While many job categories are somewhat tied to oil and gas production,
we focus on the categories that are at higher potential risk: Oil and gas extraction,
drilling of oil and gas wells, support activities 3, and petroleum refineries. The share
of such jobs is more than 5 percent of total employment in Wyoming, North
Dakota, and Alaska, putting those areas at higher risk. Meanwhile, total rig count
across the United States dropped from 1,925 in October 2014 to 885 in May 2015 (as
of May 22), a 54 percent reduction. Rig counts declined in all regions except Alaska,
with some states experiencing nearly a 50 percent reduction. This is likely evidence
that low oil prices are having a negative impact on drilling activities, which in turn
will affect future production and job growth.
In fact, the labor market has already started to feel the effects of the low oil prices.
According to Challenger, Gray & Christmas, Inc., job cuts announced in first quarter
2015 were up 16 percent, to 140,000, compared to first quarter 2014. Of these cuts,
34 percent were directly related to falling oil prices. Furthermore, the majority of
these cuts were in the energy sector, at 38,000, and Texas reported the largest
number of overall job cuts at 47,000. While the pace of energy job cuts slowed in
March compared to January and February, cuts in industrial goods picked up in
March as manufactures are adjusting their payrolls accordingly.
Exhibit 2 – Oil and Gas Job Concentration, Oil Rig Count and Recent Drop in Oil
Rigs
State
Wyoming
North Dakota
Alaska
Oklahoma
Louisiana
New Mexico
Texas
Colorado
West Virginia
Montana
California
Subtotal
United States
Oil & Gas Job
Concentration
(State)
8.4%
7.2%
5.6%
4.7%
3.7%
3.6%
3.2%
1.4%
1.3%
1.3%
0.3%
0.5%
0.6%
Percentage of Rig
Count to Total U.S.
Rigs*
2.5%
8.8%
1.0%
11.8%
7.8%
5.3%
42.1%
4.4%
2.0%
0.0%
1.5%
87%
100%
Total Rigs as
of October
2014
61
182
8
208
111
100
899
76
33
11
45
1736
1925
Drop in Rig
Count Since
October 2014**
-39
-104
1
-104
-42
-53
-526
-37
-15
-11
-32
-964
-1040
Source: Bureau of Labor Statistics, Baker Hughes, Freddie Mac
*As of May 22, 2015
**As of May 22, 2015, compared to the month average for October 2014
3
Support activities include jobs in support of the oil- and gas-extraction process that can be performed on a contract
or fee basis or by mining operators’ in-house staff.
4
Higher-risk Metros
Smaller MSAs with a
high dependence on oil
and gas jobs are exposed
to the most risk due to
less diversified
economies.
Next, we narrow down the analyses by MSA (Metropolitan Statistical Area). Metros
around the oil patch regions are typically smaller. However, several larger MSAs have
a relatively high concentration of oil and gas jobs. Exhibit 3 lists the metros with a
relatively high concentration of oil and gas jobs and their industrial diversification
indexes. The industrial diversification index, measured by Moody’s Analytics, shows
how closely the area’s economy resembles the diversification of the entire U.S.
economy. The closer to one, the better diversified; the closer to zero, the less
diversified. The exhibit contrasts larger metros with smaller ones. Notably, smaller
metros generally have a higher concentration of oil- and gas-related jobs.
Furthermore, oil- and gas-related job industries are the dominant driver of the local
economies in these metros, as evidenced by the lower industrial diversification
values. As such, smaller metros possess higher risk that the local economy will not
be able to fully absorb the job losses in the event of layoffs in the oil and gas
industries. Of these, the highest risk is in Midland and Odessa, both located within
the Permian Basin in Texas. These metros have the largest concentration of oil and
gas jobs -- 15.7 and 12.3 percent, respectively -- and very low industrial
diversification index ratings, 0.08 and 0.14, respectively.
Unlike in smaller MSAs, the share of oil and gas jobs in larger metros is relatively
small, and these metros generally are more diversified. However, the absolute
number of oil and gas jobs in these metros is far greater than in smaller metros. For
example, Houston, the epicenter of the oil and gas industry, houses almost 50
percent of all oil and gas jobs in Texas; however, these jobs compose just 4.8 percent
of the metro’s total employment. Houston’s industrial diversity index is 0.59, which
is the same as Los Angeles and New York City, an indication that the employmnet
market is well diversified. As such, the metro will be able to withstand oil- and gasrelated downturns more effectively than many of the smaller metros. But compared
to other larger metros, Houston possesses the highest potential risk. Oklahoma City,
for example, with a 4.2 percent share of oil and gas jobs, has a higher diversity index
of 0.70. Other larger metros have a much lower share of jobs in oil and gas industries
and higher diversification values.
5
Exhibit 3 – Oil and Gas Concentration and Industrial Diversification Index by MSA
Metro Area
Oil & Gas Job
Concentration
Industrial
Diversification
Index
Larger MSAs
Metro Area
Oil & Gas
Industrial
Job
Diversification
Concentration Index
Smaller MSAs
Houston, TX
Oklahoma City, OK
New Orleans, LA
Tulsa, OK
Dallas, TX
Denver, CO
San Antonio, TX
Austin, TX
United States
4.8%
4.2%
2.5%
1.9%
1.1%
1.0%
0.9%
0.4%
0.6%
0.59
0.70
0.61
0.62
0.80
0.80
0.81
0.67
1.00
Midland, TX
Odessa, TX
Casper, WY
Lafayette, LA
Greeley, CO
Farmington, NM
Houma, LA
Grand Junction, CO
15.7%
12.3%
11.0%
9.8%
8.8%
7.9%
6.7%
5.6%
0.08
0.14
0.25
0.20
0.37
0.21
0.07
0.48
Source: U.S. Bureau of Labor Statistics, Moody’s Analytics, Freddie Mac
It is important to note that not all of the oil patch regions are part of an MSA. This
limits the ability to narrow in on where to assess the greatest risk exposure in these
regions. Many of the oil fields in states such as Wyoming and North Dakota are
sparsely populated and very few or no MSAs in those states show up in the risk
analysis. We expect any such market to have a very high concentration of jobs in oil
and gas industries and very low economic diversification.
What Others See
To fully understand the potential impacts to the U.S. economy, a large-scale
structural model could be used to capture the direct and indirect effects of specific
scenarios. Moody’s Analytics developed such a model, which captures the
interactions between economic, financial, and demographic drivers in the economy.
Given this tool, Moody’s Analytics created a low-oil-price scenario. This scenario
assumes oil will stay around $60 per barrel until the end of 2017, compared to its
baseline scenario that assumes oil prices will increase steadily over the next three
years to above $100 by the end of 2017.
Even if oil prices remain
low for the next three
years, Moody’s
Analytics predicts
national employment
growth will not be
significantly impacted
and may even improve.
In general, the Moody’s forecasts are intuitive. National employment growth is
expected to remain consistent between the low-oil-price and baseline scenarios in the
short run, around 2.5 percent. But over the next three years, the employment growth
forecast in the low-oil-price scenario is stronger than in the baseline scenario because
of the cost-savings boost in oil-consuming industries.
Many energy-dependent areas, both states and metros, are expected to see an impact
on job growth over the next few years compared to the baseline scenario. The larger,
more diversified metros are likely to absorb some of the job losses from the oil and
gas industries and experience a slowdown in employment. However, in smaller
metros where the job market is less diversified and the share of oil and gas jobs is
6
high, total employment is expected to decrease under the low-oil-price scenario. The
impact is expected to be worst in 2015 before employment turns the corner in 2016
and 2017.
Impact on the Multifamily Rental Housing Sector
Housing preferences vary greatly among these areas; smaller areas are more singlefamily oriented, while larger MSAs have a greater concentration of multifamily
properties. Nonetheless, several smaller areas have seen a spike in multifamily and
single-family construction in recent years due to the large influx of workers.
Areas with a recent
increase in multifamily
supply are at higher risk
of experiencing
unfavorable multifamily
fundamentals.
Multifamily construction has been accelerating from historically low levels in most
metros in the post-recession years. Nationwide, multifamily construction starts
ended 2014 at 341,000 units, about 50,000 units higher than the annual long-run
average from 1995-2007. The growth is largely driven by the strengthening economy
and lagging single-family market. However, the energy boom has spurred even
higher growth in many of the areas with high concentration of oil and gas jobs.
Exhibit 4 shows the percentage of multifamily housing starts in 2014 compared to
the historical average in some of these states and MSAs. North Dakota and Houston
have seen a significant rise in multifamily housing construction over the last few
years to accommodate the rise in employment. If employment slows over the next
few years and the amount of building exceeds the demand, multifamily fundamentals
could be strained. The effects are not expected to be as severe in the other states and
MSAs, but lower employment growth over the next year could lessen multifamily
performance expectations in some submarkets. However, while the multifamily
sector in smaller areas is at risk as well, these areas contain much less multifamily
housing stock than in relatively larger MSAs.
Exhibit 4 – Multifamily Starts in 2014 Compared to Historical Average by State and
MSA
State
Wyoming
North Dakota
Alaska
Oklahoma
Louisiana
New Mexico
Texas
Colorado
California
Multifamily Starts in
2014 Compared to
Historical Average*
16%
495%
-52%
37%
-18%
-21%
67%
15%
26%
MSA
Houston
Oklahoma City
New Orleans
Tulsa
Dallas
Denver
San Antonio
Austin
United States
Multifamily Starts in
2014 Compared to
Historical Average*
90%
37%
-42%
77%
38%
21%
-36%
49%
7%
*Historical average calculated as the average starts from 1995-2007
Source: U.S. Census Bureau, Moody’s Analytics, Freddie Mac
7
Section 2 – Stress Scenario: Houston
Persistently low oil prices will have an impact on highly exposed markets. Our
analysis focuses on the Houston market because of its relatively heavy dependence
on the energy sector and recent boom in multifamily construction, as discussed in
the prior section. While today’s economy is different than in the 1980s, measuring
the potential impact low oil prices have on the market is desirable. We create a stress
scenario to better understand current conditions in the Houston market and how
that compares to intense strain on the market. In our stress scenario, we determine
what level of impact is needed to create a negative shock in the Houston economy
and evaluate its effects on the multifamily market. We find that several factors in the
analysis contrast current market conditions, and conclude that the market is not on
the edge of a precipitous decline.
Impact of the 1980s Oil Crash
While Houston’s multifamily sector has been through several booms and busts in the
past, the energy bust of the 1980s was the most severe. Prior to the sharp decline in
oil prices in the early 1980s, Houston’s economy was thriving. The unemployment
rate was 4.1 percent, more than 2.5 percentage points below the U.S. average, and
employment had been growing on average 7.5 percent per year, compared to 2.6
percent nationally. Because the market was attracting many workers, Houston was
the fastest growing major MSA, with 4.6 percent population growth per year on
average compared to 1 percent nationally. Both single-family and multifamily sectors
benefitted from the growing demand; construction in both housing segments
boomed, further fueling the market growth. The growing demand produced
accelerated rent growth.
After this exuberant period of growth, oil prices began to fall in 1981. Prices
dropped from $115/barrel in April 1980 to $64/barrel in November 1985, but then
dropped drastically and quickly, hitting $26/barrel by February 1986. While the oil
production and associated industries felt the pain first, the spillover effects also
impacted other sectors of the market. The effect was devastating. The total labor
market shrank by 1.7 percent annually; the unemployment rate more than doubled,
reaching 11 percent; and population growth significantly slowed to nearly 0.5 percent
annually.
The deteriorating fundamentals also took a toll on the housing market. Single-family
house price declined by 25 percent from peak to trough, while construction
plummeted by 80 percent. Multifamily construction was almost completely halted;
vacancy rates sharply rose, reaching 18 percent; and rents declined for four straight
years, dropping 14 percent in total.
8
The Energy and Multifamily Sectors in the 1980s
While the extremely strong local economy drove multifamily construction in
Houston leading into the last oil price-induced stress, it also coincided with a boom
and bust in the multifamily sector across the nation. The entire U.S. multifamily
sector grew tremendously in late 1970s to mid-1980s, largely driven by the tax shelter
that many investors benefitted from by investing in multifamily properties. In fact,
investors moved capital into commercial real estate when oil prices began to fall in
an effort to maintain high returns on invested capital. The heavy investment in
commercial real estate at the time stimulated multifamily construction across the
nation. Multifamily starts with five or more units peaked in 1985 at 577,000 units,
from around 300,000 in the early 1980s.
In comparison, while multifamily starts have been on the rise in the past few years,
they have just surpassed 300,000 in 2014 after remaining significantly below the longrun average of 295,000 from 2009 to 2012. Therefore, the multifamily construction
boom since the Great Recession is still nowhere near the boom seen in the mid1980s.
The energy crash of the
1980s also coincided
with a crash in the
multifamily sector.
Multifamily markets
enter this potential stress
much more balanced
today.
The shocks of the low oil prices and oversupply of multifamily construction in the
1980s hit markets hard in areas heavily dependent on the oil and gas industries, such
as Houston, Austin, and Dallas. Deteriorations in labor markets and property
markets resonated into fundamentals of the multifamily market, leading to loan
defaults. Losses in commercial lending for investments in the energy sector
devastated southwest regional capital markets. From 1980 through 1989, 425 Texas
commercial banks failed, including nine out of the state’s 10 largest bank holding
companies. The collapse of the capital markets resulted in the meltdown of the entire
multifamily sector. While it is difficult to isolate the net effect that oil prices had on
the capital markets, it is clear that the combination of many factors made it much
worse for energy-dependent areas.
More Diversified Economy and Relatively Low Supply
The labor markets of today are different than they were in the 1980s. According to
the Bureau of Labor Statistics’ Current Employment Statistics program, education
and health services, professional and business services, and leisure and hospitality
sectors are increasingly becoming major industries in Houston. The combined share
of employment in these sectors has increased from 28.5 percent in 1990 (the earliest
date available) to 37.4 percent in 2014. As mentioned earlier, Moody’s Analytics’
diversity index also shows that Houston economy is relatively diversified, and is
more so now than it was in the 1980s; an index of 0.59 today compared to 0.51 in
1983. Better diversification will help Houston weather shocks in the energy sector.
9
Furthermore, the supply of new multifamily units while high is not nearly as elevated
as it was in 1980s. The ratio of new construction starts relative to existing inventory
in the 1980s and now shows a dramatic difference. Multifamily starts in 1982 and
1983 comprised 15 percent of the existing multifamily inventory, in aggregate. The
share of starts in 2013 and 2014 comprised only 5 percent, in aggregate. Moreover, a
recent decline in multifamily permits is an indication of slowing multifamily
construction.
Stress Scenario
Moody’s low-oil-price
scenario anticipates flat
employment growth in
Houston in 2015;
growth in other sectors
will help offset the energy
sector job losses.
According to Moody’s Analytics’ baseline scenario employment forecast, Houston’s
labor market is likely to continue to expand at an average rate of 2.4 percent annually
for the next three years. However, in the low-oil-price scenario, employment growth
is likely to slow to 2.2 percent on average per year. The labor market impact will be
most substantial in 2015 when the stress scenario projects near zero growth; faster
growth is expected in consecutive years. It is expected that the job gains in other
sectors as a consequence of low oil prices will help offset the jobs lost in the energy
sector.
Based on these forecasts, multifamily fundamentals are likely to slow in the shortterm, but will not have a major, long-lasting, metro-wide impact. Nevertheless, we
consider multiple scenarios for the factors affecting the multifamily market. We use
our internal market forecast model to assess the impact of these paths on Houston’s
rent growth and vacancy rate. The model includes functionality that considers
historical market “turning points”. Building scenarios consistent with previously
identified conditions driving these dramatic changes in the market can inform us
about current conditions. Shocks are likely to occur when major disruptions take
place in the supply side, demand side, or both. For example, a turning point occurred
in the 1980s when demand quickly fell and the large amount of new supply entered
the market, exacerbating the market collapse.
Our baseline scenario is in line with Moody’s Analytics’ baseline scenario; it assumes
that Houston employment will continue to grow at moderate levels, albeit a slower
rate than in the past few years. It also assumes that multifamily construction will
gradually slow. Rent growth is expected to gradually slow and vacancy rates to inch
up to 7.2 percent. Even when we include the path with employment growth and
multifamily starts based on Moody’s low-oil-price scenario, there was little impact to
our forecast of multifamily fundamentals.
However, we implement a stress path scenario to force the Houston market to
experience a turning point. This scenario assumes that employment will decline 4
percent and that multifamily construction will keep increasing to near 40,000 units
annually. These extreme conditions are not forecast to occur, making it unlikely the
10
market will shift as projected. Vacancy rates in this scenario will spike up to 10
percent by 2016 and remain flat in 2017, while rents will decline 1.1 percent in 2015
and continue to slide another 3.7 percent in 2016 before turning positive again in
2017. It is imperative to note that the probability of such a scenario is very low
because, as explained earlier, the growth in employment is likely to slow down, not
turn negative, and multifamily development is likely to decline instead of increase. As
shown in Exhibit 5, the effect of such a shock is more like that of the Great
Recession than of the energy crisis of 1980s.
Exhibit 5 – Houston Historical and Projected Vacancy Rates and Market Rent
Growth
Source: REIS, Freddie Mac
Conclusion
Low oil prices likely will be a net benefit to the overall economy, although smaller
markets with a high dependency on energy-related jobs are expected to be negatively
impacted. A few larger metros with a significant concentration of energy jobs could
see an impact to their economies and multifamily performance over the next few
years. However, in these larger areas the severity of the downturn will be less and the
market will rebound more quickly.
In Houston, even if oil prices remain depressed for three years, the shock to the
economy and labor market will not be as stressful as the one experienced in the
1980s. Furthermore, in order to shock the Houston economy so that multifamily
fundamentals are impacted severely, the drop in employment would have to be
consistent with that of the Great Recession and multifamily starts concurrently
would have to increase significantly. Even then, the impact in this type of scenario
does not produce the severity of the 1980s oil price shock.
11
Instead, current expectations are that employment growth will slow and multifamily
construction will follow suit. Some areas in Houston will be impacted, especially
those where new multifamily deliveries are still in lease-up. But with employment
growth expected to rebound in 2016 and beyond, impacts to the rental market are
not expected to be severe. However, smaller, less diversified markets with heavy ties
to the energy sector may experience more severe impacts to their economy and
housing markets.
For more insights from the Freddie Mac Multifamily Research team, visit the Research page on
FreddieMac.com/Multifamily.
12